Seedance 2.0 Video API: Text and Image to Video

Available models

Loading…

See video models in the Model Catalog and filter for Video. Charges are actual duration multiplied by the model price.

Two-step asynchronous flow

Submit: POST /v1/videos immediately returns an id with queued status.
Poll GET /v1/videos/{id} every few seconds until completed or failed. A five-second Seedance video usually takes about one to two minutes.

Submission reserves the estimated cost; success settles the actual duration and model price idempotently. Failure or timeout releases the reservation, and repeated polling never charges twice.

Submission parameters

Body parametersapplication/json

modelstringRequired

A Seedance model. See available models above.

Example:your-seedance-model-id

promptstringRequired

A description of the scene. Chinese prompts are supported.

Example:An orange cat running through a sunny field, cinematic lighting

durationintegerOptional

Duration in seconds. Supported values depend on the model; Seedance commonly supports 5 or 10 seconds.

Example:5

resolutionstringOptional

Output resolution such as 720P or 1080P.

Example:720P

aspect_ratiostringOptional

Aspect ratio such as 16:9, 9:16, or 1:1.

Example:16:9

bitrate_modestringOptional

Seedance output bitrate mode. Supported values are standard and high; omitted values use the model default, then the upstream default when unconfigured.

Example:standard

negative_promptstringOptional

An optional negative prompt.

Example:blur,distortion

seedintegerOptional

Random seed. -1 selects a random seed.

Example:-1

Volcano-specific behavior

Returns a temporary MP4 link from Volcano TOS that usually expires after about 24 hours. Download it promptly.
Only successful tasks are billed by duration; for example, a five-second video costs 5 × the per-second price.
Supports text-to-video, image-to-video, and first/last-frame interpolation.

Image to video / first and last frames

Add an image URL or base64 data URI to the request body for image-to-video without uploading a file.

Additional body parameters for image-to-video

imagestring | string[]Optional

Reference images. One image becomes the first frame; multiple images are used as reference material.

Example:https://...png

imagesstring[]Optional

Multiple reference images. Array order defines image 1, image 2, and so on. Public URLs and data URIs are accepted.

Example:["https://.../person.png","https://.../scene.png"]

first_framestringOptional

The first keyframe, optionally combined with last_frame for interpolation.

Example:https://...png

last_framestringOptional

The final keyframe.

Example:https://...png

Image to video · curl

# Image to video: pass an image URL (or base64 data URI) in image to animate it
curl https://zerofa.ai/v1/videos \
  -H "Authorization: Bearer sk-zerofa-xxx" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "your-seedance-model-id",
    "prompt": "The cat stands and stretches as the camera slowly pushes in",
    "image": "https://your-cdn.com/cat.png",
    "duration": 5
  }'

# First/last-frame interpolation: provide both frames to generate the transition
#   "first_frame": "https://.../a.png",
#   "last_frame":  "https://.../b.png"

Multimodal references (r2v) · Seedance 2.0

Seedance 2.0 can combine image, video, and audio references to borrow camera movement, style, music, or lip-sync cues.

Additional body parameters for multimodal references

reference_videostring | string[]Optional

Reference video accepts a public URL or asset ID, not a base64 data URI; Seedance 2.0 supports up to three clips.

Example:https://...mp4

reference_audiostring | string[]Optional

Up to three reference audio clips. Audio cannot be used alone and requires at least one image or video.

Example:https://...mp3

Seedance 2.0 supports up to 9 images, 3 videos, and 3 audio clips, with a maximum duration of 15 seconds.

Multimodal references · curl

# Multimodal reference (Seedance 2.0): combine image, video, and audio references
curl https://zerofa.ai/v1/videos \
  -H "Authorization: Bearer sk-zerofa-xxx" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "your-seedance-model-id",
    "prompt": "Use this camera movement with a dog running on a beach as the subject",
    "images": ["https://.../dog.png"],
    "reference_video": ["https://.../style.mp4"],
    "reference_audio": ["https://.../bgm.mp3"],
    "duration": 5
  }'

Retrieve a task

GET/v1/videos/{id}

The response is an OpenAI Videos-compatible object with one of four statuses: queued, in_progress, completed, or failed. Terminal results are cached, so repeated retrieval does not call the upstream again. progress is a stage indicator, not frame-level upstream progress.

Request and response bodies

Use the examples below to verify the request and response structures. To send a request, select Debug at the top of the page.

① Submit

curl https://zerofa.ai/v1/videos \
  -H "Authorization: Bearer sk-zerofa-xxx" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "your-seedance-model-id",
    "prompt": "An orange cat running through a sunny field, cinematic lighting",
    "duration": 5,
    "resolution": "720P",
    "aspect_ratio": "16:9",
    "bitrate_mode": "standard"
  }'
# → {"id":"<task-id>","status":"queued"}

② Poll

curl https://zerofa.ai/v1/videos/<task-id> \
  -H "Authorization: Bearer sk-zerofa-xxx"

# in_progress → {"id":"...","status":"in_progress"}
# completed   → {"id":"...","status":"completed",
#                "metadata":{"url":"https://.../result.mp4"}}

Two-step asynchronous flow

Submit: POST /v1/videos immediately returns an id with queued status.
Poll GET /v1/videos/{id} every few seconds until completed or failed. A five-second Seedance video usually takes about one to two minutes.

Submission reserves the estimated cost; success settles the actual duration and model price idempotently. Failure or timeout releases the reservation, and repeated polling never charges twice.

Submission parameters

Body parametersapplication/json

modelstringRequired

A Seedance model. See available models above.

Example:your-seedance-model-id

promptstringRequired

A description of the scene. Chinese prompts are supported.

Example:An orange cat running through a sunny field, cinematic lighting

durationintegerOptional

Duration in seconds. Supported values depend on the model; Seedance commonly supports 5 or 10 seconds.

Example:5

resolutionstringOptional

Output resolution such as 720P or 1080P.

Example:720P

aspect_ratiostringOptional

Aspect ratio such as 16:9, 9:16, or 1:1.

Example:16:9

bitrate_modestringOptional

Seedance output bitrate mode. Supported values are standard and high; omitted values use the model default, then the upstream default when unconfigured.

Example:standard

negative_promptstringOptional

An optional negative prompt.

Example:blur,distortion

seedintegerOptional

Random seed. -1 selects a random seed.

Example:-1

Image to video / first and last frames

Add an image URL or base64 data URI to the request body for image-to-video without uploading a file.

Additional body parameters for image-to-video

imagestring | string[]Optional

Reference images. One image becomes the first frame; multiple images are used as reference material.

Example:https://...png

imagesstring[]Optional

Multiple reference images. Array order defines image 1, image 2, and so on. Public URLs and data URIs are accepted.

Example:["https://.../person.png","https://.../scene.png"]

first_framestringOptional

The first keyframe, optionally combined with last_frame for interpolation.

Example:https://...png

last_framestringOptional

The final keyframe.

Example:https://...png

Image to video · curl

# Image to video: pass an image URL (or base64 data URI) in image to animate it
curl https://zerofa.ai/v1/videos \
  -H "Authorization: Bearer sk-zerofa-xxx" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "your-seedance-model-id",
    "prompt": "The cat stands and stretches as the camera slowly pushes in",
    "image": "https://your-cdn.com/cat.png",
    "duration": 5
  }'

# First/last-frame interpolation: provide both frames to generate the transition
#   "first_frame": "https://.../a.png",
#   "last_frame":  "https://.../b.png"

Multimodal references (r2v) · Seedance 2.0

Seedance 2.0 can combine image, video, and audio references to borrow camera movement, style, music, or lip-sync cues.

Additional body parameters for multimodal references

reference_videostring | string[]Optional

Reference video accepts a public URL or asset ID, not a base64 data URI; Seedance 2.0 supports up to three clips.

Example:https://...mp4

reference_audiostring | string[]Optional

Up to three reference audio clips. Audio cannot be used alone and requires at least one image or video.

Example:https://...mp3

Seedance 2.0 supports up to 9 images, 3 videos, and 3 audio clips, with a maximum duration of 15 seconds.

Multimodal references · curl

# Multimodal reference (Seedance 2.0): combine image, video, and audio references
curl https://zerofa.ai/v1/videos \
  -H "Authorization: Bearer sk-zerofa-xxx" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "your-seedance-model-id",
    "prompt": "Use this camera movement with a dog running on a beach as the subject",
    "images": ["https://.../dog.png"],
    "reference_video": ["https://.../style.mp4"],
    "reference_audio": ["https://.../bgm.mp3"],
    "duration": 5
  }'

Request and response bodies

Use the examples below to verify the request and response structures. To send a request, select Debug at the top of the page.

① Submit

curl https://zerofa.ai/v1/videos \
  -H "Authorization: Bearer sk-zerofa-xxx" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "your-seedance-model-id",
    "prompt": "An orange cat running through a sunny field, cinematic lighting",
    "duration": 5,
    "resolution": "720P",
    "aspect_ratio": "16:9",
    "bitrate_mode": "standard"
  }'
# → {"id":"<task-id>","status":"queued"}

② Poll

curl https://zerofa.ai/v1/videos/<task-id> \
  -H "Authorization: Bearer sk-zerofa-xxx"

# in_progress → {"id":"...","status":"in_progress"}
# completed   → {"id":"...","status":"completed",
#                "metadata":{"url":"https://.../result.mp4"}}

Volcano Engine video (Seedance 2.0)

Available models

Two-step asynchronous flow

Submission parameters

Volcano-specific behavior

Image to video / first and last frames

Multimodal references (r2v) · Seedance 2.0

Retrieve a task

Request and response bodies

Volcano Engine video (Seedance 2.0)

Available models

Two-step asynchronous flow

Submission parameters

Volcano-specific behavior

Image to video / first and last frames

Multimodal references (r2v) · Seedance 2.0

Retrieve a task

Request and response bodies